The Underside of Global Warming: Ojai Valley Cooling
Olivia Whitener
2/6/2017
## [1] "19170101" "19170102" "19170103" "19170104" "19170105" "19170106"## [1] OJAI CA US
## Levels: OJAI CA US##
## Call:
## lm(formula = TMAX ~ NewDate, data = LosAngeles)
##
## Residuals:
## Min 1Q Median 3Q Max
## -44.919 -9.415 -0.342 9.350 40.416
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.807e+01 6.640e-02 1175.888 < 2e-16 ***
## NewDate -2.658e-05 6.259e-06 -4.247 2.17e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 12.41 on 35529 degrees of freedom
## (782 observations deleted due to missingness)
## Multiple R-squared: 0.0005074, Adjusted R-squared: 0.0004793
## F-statistic: 18.04 on 1 and 35529 DF, p-value: 2.172e-05
## Month Year TMAX YEAR MONTH NewDate
## 1 01 1917 7.069547 1917 1 1917.000
## 2 02 1917 8.784908 1917 2 1917.083
## 3 03 1917 7.809974 1917 3 1917.167
## 4 04 1917 8.435284 1917 4 1917.250
## 5 05 1917 6.535758 1917 5 1917.333
## 6 06 1917 12.658194 1917 6 1917.417
##
## Call:
## lm(formula = TMAX ~ NewDate, data = MonthlySD)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.5948 -1.3109 -0.1002 1.2044 7.9359
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 10.564453 3.578222 2.952 0.00322 **
## NewDate -0.001456 0.001819 -0.800 0.42385
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.802 on 1182 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0005412, Adjusted R-squared: -0.0003043
## F-statistic: 0.6401 on 1 and 1182 DF, p-value: 0.4238
For the past few decades, the scientific community has been locked in a fierce debate over the issue of climate change. Those concerned about the future of the environment, known as “alarmists,” warn that increased carbon dioxide emissions and the resulting global temperature increase of about 1.7 degrees Fahrenheit since 1880 (NASA) will have disastrous effects on agriculture, ecosystems, water resources, and human health. Although “skeptics” may acknowledge increased temperatures, they deny that there is anything out of the ordinary that should cause humans to change behavior.
The Golden State of California, already known for its sunny skies and breezy coasts, has been warming for the past century. Southern California has experienced a dramatic temperature increase of 3 degrees Fahrenheit in the past century. With increased temperatures come the risks of heat waves, drought, early snow melt, and decreased precipitation (EPA). If this is the case, then why have the maximum temperatures of more than half the year in Ojai, California decreased since 1920?
Ojai is a small town of about 7500 people situated in Ventura County, just northwest of Los Angeles and east of Santa Barbara. According to National Oceanic and Atmospheric Association climate data, the maximum temperatures in Ojai have decreased for seven out of twelve months. Five of these months, four of which have decreasing temperatures, have p-values less than 0.05. Therefore this data is statistically significant, which means that it rejects the null hypothesis that modern climate change is due to natural causes.
Selecting for 1 Month
##
## Call:
## lm(formula = TMAX ~ YEAR, data = MonthlyMean[MonthlyMean$Month ==
## "04", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -10.1433 -2.5047 -0.0938 2.8245 8.0180
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 21.6564 25.9442 0.835 0.4059
## YEAR 0.0267 0.0132 2.023 0.0458 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.75 on 96 degrees of freedom
## Multiple R-squared: 0.0409, Adjusted R-squared: 0.03091
## F-statistic: 4.094 on 1 and 96 DF, p-value: 0.04582
April is the only statistically significant month that is increasing in temperature.
##
## Call:
## lm(formula = TMAX ~ YEAR, data = MonthlyMean[MonthlyMean$Month ==
## "07", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.9677 -2.0630 -0.1102 2.3523 8.7120
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 191.52739 22.20627 8.625 1.24e-13 ***
## YEAR -0.05114 0.01129 -4.529 1.69e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.238 on 97 degrees of freedom
## Multiple R-squared: 0.1745, Adjusted R-squared: 0.166
## F-statistic: 20.51 on 1 and 97 DF, p-value: 1.69e-05
##
## Call:
## lm(formula = TMAX ~ YEAR, data = MonthlyMean[MonthlyMean$Month ==
## "08", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.8232 -1.5971 -0.0065 1.5255 8.5973
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 134.19879 20.15496 6.658 1.61e-09 ***
## YEAR -0.02162 0.01025 -2.109 0.0375 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.958 on 98 degrees of freedom
## Multiple R-squared: 0.04343, Adjusted R-squared: 0.03367
## F-statistic: 4.449 on 1 and 98 DF, p-value: 0.03746
##
## Call:
## lm(formula = TMAX ~ YEAR, data = MonthlyMean[MonthlyMean$Month ==
## "11", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.8244 -3.3259 0.3259 2.7188 10.2763
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 163.69455 27.98160 5.85 6.65e-08 ***
## YEAR -0.04539 0.01423 -3.19 0.00192 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.08 on 97 degrees of freedom
## Multiple R-squared: 0.09493, Adjusted R-squared: 0.0856
## F-statistic: 10.17 on 1 and 97 DF, p-value: 0.001919
##
## Call:
## lm(formula = TMAX ~ YEAR, data = MonthlyMean[MonthlyMean$Month ==
## "12", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.9694 -3.1011 -0.3825 3.0747 9.2401
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 137.85146 27.48570 5.015 2.42e-06 ***
## YEAR -0.03572 0.01398 -2.555 0.0122 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.958 on 96 degrees of freedom
## Multiple R-squared: 0.06365, Adjusted R-squared: 0.0539
## F-statistic: 6.526 on 1 and 96 DF, p-value: 0.0122
The 2006 California Climate Change Center report projected warming from 2000 to 2100 to vary from approximately 3.0°F–5.4°F in the lower range of projected warming to 8.0°F–10.4°F in the higher range. Higher temperatures inland create pressure differences that affect wind patterns. These winds can bring hotter, dryer desert air to the Southwest, change the cloud cover over the Pacific Northwest, or shift rain patterns in the Southern US. They also bring cooler air from the ocean, lowering the temperatures of coastal regions such as Ojai.
TMIN
- We create a monthly mean for each month.
## Month Year TMIN YEAR
## 1 01 1917 34.58065 1917
## 2 02 1917 35.92857 1917
## 3 03 1917 33.35484 1917
## 4 04 1917 39.46667 1917
## 5 05 1917 41.09677 1917
## 6 06 1917 51.60000 1917- Now we plot the mins, and again, find tons of scatter.
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "01", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.5486 -1.7755 -0.1755 1.3592 7.8665
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -26.78993 19.92827 -1.344 0.18195
## YEAR 0.03183 0.01013 3.142 0.00222 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.948 on 98 degrees of freedom
## Multiple R-squared: 0.09151, Adjusted R-squared: 0.08224
## F-statistic: 9.871 on 1 and 98 DF, p-value: 0.00222
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "03", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.073 -1.373 -0.519 1.671 6.960
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -21.363213 17.456061 -1.224 0.224012
## YEAR 0.031190 0.008879 3.513 0.000678 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.523 on 96 degrees of freedom
## Multiple R-squared: 0.1139, Adjusted R-squared: 0.1047
## F-statistic: 12.34 on 1 and 96 DF, p-value: 0.000678
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "05", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.6486 -1.6267 0.0623 1.6795 6.4054
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -5.887086 16.532669 -0.356 0.72256
## YEAR 0.027011 0.008409 3.212 0.00179 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.39 on 96 degrees of freedom
## Multiple R-squared: 0.09704, Adjusted R-squared: 0.08764
## F-statistic: 10.32 on 1 and 96 DF, p-value: 0.001794
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "06", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.3810 -1.2987 0.0021 1.4953 5.1190
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -23.090329 14.389683 -1.605 0.112
## YEAR 0.037559 0.007318 5.132 1.47e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.095 on 97 degrees of freedom
## Multiple R-squared: 0.2136, Adjusted R-squared: 0.2055
## F-statistic: 26.34 on 1 and 97 DF, p-value: 1.47e-06
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "07", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.7796 -1.6199 -0.1003 1.7751 8.3318
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -14.601690 17.259527 -0.846 0.399631
## YEAR 0.035276 0.008777 4.019 0.000116 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.516 on 97 degrees of freedom
## Multiple R-squared: 0.1427, Adjusted R-squared: 0.1339
## F-statistic: 16.15 on 1 and 97 DF, p-value: 0.0001155
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "08", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.8051 -2.0273 -0.2833 2.0279 7.7257
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -10.458953 17.113818 -0.611 0.542521
## YEAR 0.033074 0.008702 3.801 0.000251 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.512 on 98 degrees of freedom
## Multiple R-squared: 0.1285, Adjusted R-squared: 0.1196
## F-statistic: 14.45 on 1 and 98 DF, p-value: 0.0002506
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "09", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.1896 -1.8215 0.0521 1.6751 9.3185
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -24.61825 18.66327 -1.319 0.19
## YEAR 0.03916 0.00949 4.127 7.72e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.739 on 98 degrees of freedom
## Multiple R-squared: 0.1481, Adjusted R-squared: 0.1394
## F-statistic: 17.03 on 1 and 98 DF, p-value: 7.723e-05
##
## Call:
## lm(formula = TMIN ~ YEAR, data = MonthlyMeanTMIN[MonthlyMeanTMIN$Month ==
## "10", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.1782 -1.7592 -0.3477 1.8006 6.4333
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -9.124970 18.153567 -0.503 0.61635
## YEAR 0.028483 0.009232 3.085 0.00265 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.647 on 97 degrees of freedom
## Multiple R-squared: 0.08936, Adjusted R-squared: 0.07997
## F-statistic: 9.518 on 1 and 97 DF, p-value: 0.002651
Experimental Portion — Precipitation
Precipiation might depend more on the departure from the mean (often referred as as normal, whatever that means!). I think it’s worth pursuing, but haven’t finished the analysis yet.
First, we need a “mean” – The IPCC uses 1961-1990 as a norm, I don’t know what is the standard for California, so we should look that up.
Second, we need to remove the missing values and evalaute which years have complete years. If you are missing rainy months, then the whole year should be thrown out – but what about partial years in the drought season?
Third, we will need to decide what level of aggredation – monthly, yearly, etc.
Fourth, in CA the water year starts in Oct 1. Should we follow the same convention?
A yearly mean, based on the annual sum for the entire records. Not sure this is appropriate.
Figure has points of the yearly sum of rainfall and the blue line mean. The greenline is the trend and red line is a five year running average, I think! I am still trying to understand what the code is doing.
## [1] 0.2 0.2 0.2 0.2 0.2
The model suggests that the precipitation is declines at a rate of -0.0119165 cm yr\(^{-1}~\), or -0.12 cm decade\(^{-1}\).
##
## Call:
## lm(formula = PRCP ~ YEAR, data = YearlySum)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.517 -6.829 -2.899 5.132 27.802
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 43.06833 67.06418 0.642 0.522
## YEAR -0.01192 0.03409 -0.350 0.727
##
## Residual standard error: 9.989 on 99 degrees of freedom
## Multiple R-squared: 0.001233, Adjusted R-squared: -0.008856
## F-statistic: 0.1222 on 1 and 99 DF, p-value: 0.7274